Network operators frequently desire to redirect Internet Protocol (IP) traffic passing through its network. Some network operators may desire to be able to route traffic, especially (but not limited to) Hypertext Transfer Protocol (HTTP) traffic, through an intermediate server based on the domain name for which the IP traffic is destined. However, problems arise because the domain name layer is contained at the application layer and network load balancers and routers do not have access to the domain name as they investigate data packets at the IP layer. For a service (like a load balancer or router) to obtain the domain name, the service would likely have to perform some packet inspection, which is a costly operation in terms of network performance.
One potential solution for this problem may be to use a domain name system (DNS) server in the network to redirect the traffic. However, conventional DNS redirecting methods encounter problems when trying to redirect domain names for which the DNS is authoritative but that are not recognized by the DNS server. One such problem occurs when the DNS server blocks domain names for which the DNS server should be authoritative, but is not. This can be prevented by adding all resource records for a particular domain name to the DNS server, but this solution is impracticable and can be time consuming.